Researchers at Weill Cornell Medical College in New York have discovered that a toxin made by the bacterium Clostridium perfringens — responsible for a million cases of food poisoning in the U.S. each year — appears to attack the cells associated with MS.

“What we’ve shown is the toxins target the cells that are targeted in MS,” said Jennifer Linden, a Weill Cornell researcher who’s presenting findings Tuesday at a meeting of the American Society for Microbiology.

Her research also showed that about 13 percent of a small sample of food products harbored C. perfringens bacteria, and nearly 3 percent were positive for the toxin linked to MS.

It’s too early to suggest that MS is caused by food poisoning, but it does raise the possibility that the C. perfringens bacteria could be involved in activating the disease, said Bruce F. Bebo, associate vice president of discovery research for the National Multiple Sclerosis Society.

“To me, if you were going to design a trigger for this disease, this would really fit the mold really well,” Bebo said.

Specifically, it’s the epsilon toxin, one of 12 protein toxins produced by C. perfringens bacteria. Not only does the toxin attack the brain’s blood-brain barrier, it also kills the brain’s myelin-producing cells, the same cells that die in MS lesions, Linden said.

MS is a disease in which the fatty sheath around the nerves of the brain and spinal cord, called myelin, is destroyed, leading to a wide range of symptoms.

MS is thought to be caused by an environmental trigger, something scientists have sought for years, Bebo said. But the new theory from Weill Cornell’s team led by Timothy Vartanian, a professor of neurology and neuroscience, is particularly promising, Bebo said.

Most cases of C. perfringens food poisoning in people are caused by the A strain of the bacteria, one of five types. The B and D types, which produce the epsilon toxin, have rarely been seen in people. But last year, Weill Cornell researchers reported that they’d discovered the B strain in a 21-year-old MS patient, proving infection was possible.

They also examined banked blood and spinal fluid samples from MS patients and healthy controls and found that the MS patients were 10 times more likely to show evidence of antibodies signaling they’d previously been exposed to the epsilon toxin.

If the link with MS is proven, it’s possible that a vaccine or drug or probiotic treatment could be developed to destroy the C. perfringens B and D bacteria, which make the toxin, Vartanian has said.